A managed headful Chrome browser designed for agents, featuring anti-detection, human-in-the-loop VNC takeover, and multi-session isolation.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install anemone-browser
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install anemone-browser using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Anemone Browser?
Anemone Browser provides a high-performance, headful Chrome environment specifically optimized for Openclaw Skills. By utilizing real browser windows rather than headless instances, it bypasses common automation triggers like navigator.webdriver flags and stealth-check scripts. It serves as a bridge between automated agent logic and human intervention, allowing users to jump into a session via VNC to solve CAPTCHAs or manage complex logins without losing the agent's progress.
This skill is essential for developers building robust automation workflows that need to survive on the modern web. It handles the complexities of window management, ensuring that multiple agent sessions can operate concurrently using the same authenticated profile without interfering with each other's navigation or data. With built-in support for macOS, Linux, and Docker, it standardizes how your Openclaw Skills interact with web content.
Anemone Browser Use Cases
- Automating interactions with sites protected by Cloudflare, Google, or other anti-bot services.
- Facilitating human-in-the-loop workflows where an agent needs manual assistance for 2FA or CAPTCHA solving.
- Managing multiple simultaneous web-based agent tasks through strict window isolation.
- Standardizing browser deployment for Openclaw Skills across local machines and cloud servers.
How Anemone Browser Works
- The user installs and launches the managed browser environment using platform-specific setup scripts.
- The browser starts in headful mode with specific security flags disabled to prevent detection as an automated tool.
- Openclaw agents initiate sessions by opening new windows via the Chrome DevTools Protocol (CDP), assigning each a unique targetId.
- For all navigation and scraping actions, the agent tracks its specific targetId to maintain strict isolation from other active sessions.
- If a roadblock is encountered, the skill generates a secure noVNC link (on Linux) or notifies the user to intervene directly.
- Once the human intervention is complete, the agent resumes its automated task within the same session context.
Anemone Browser Setup
To set up on macOS, execute the following commands:
bash scripts/setup-mac.sh
openclaw browser start
For Linux or Docker environments, use the standard setup and start scripts:
bash scripts/setup.sh
bash scripts/start.sh [password] [novnc_port] [cdp_port] [resolution]
Once started, Openclaw Skills will automatically detect the CDP port and begin managing windows.
Anemone Browser Data Schema & Taxonomy
The skill organizes its operation through specific identifiers and configuration profiles to maintain persistence and isolation:
| Attribute | Description |
|---|---|
| Profile Name | Defaults to 'openclaw' to share cookies and login states across windows. |
| targetId | A unique GUID for each browser tab/window to ensure session isolation. |
| CDP URL | Local debugger URL used by agents to send commands to the browser. |
| noVNC Link | A dynamically generated URL containing the IP and credentials for remote takeover. |
Anemone Browser Advanced Features
- Anti-detection engine that disables AutomationControlled features and overrides User-Agents.
- Secure VNC takeover with SSL/TLS encryption for remote troubleshooting.
- Multi-session targetId tracking to prevent cross-contamination between different Openclaw Skills instances.
- Support for custom CDP commands to manipulate the browser at the protocol level.
- Hardened security policies that block file and javascript protocol execution to protect the host environment.
SKILL.md
Loading
Related Openclaw Skills
Abstract Searcher
A tool for automatically populating BibTeX files with academic abstracts via public APIs and intelligent browser automation.
MinerU
A sophisticated document parsing tool that converts complex PDFs, Word docs, and PPTs into AI-ready Markdown with full support for formulas and tables.
Kitful AI Article Generator
An AI-driven content engine that transforms keywords into fully researched, humanized, and publication-ready long-form articles.
Friday (星期五)
A dedicated AI programming agent designed for automated code generation, security audits, and intelligent refactoring via the opencode CLI.
CapMonster Cloud CAPTCHA Solver
An API-driven tool for programmatically bypassing reCAPTCHA, hCaptcha, Turnstile, and image CAPTCHAs during browser automation.
Grok Browser
A sophisticated browser automation tool designed to query Grok AI and extract responses via Openclaw Skills and Chrome relay.
Featured*
